Seriously?
The best path forward is:
1. Detailed visual inspection under magnification. Verify component values, part numbers & orientation. This is especially important when performing mods because mods increase the number of opportunities for mistakes.
Check all solder joints to make sure they are shiny and there are no solder shorts between traces.
Prior to inspection, clean the board to remove excess solder flux that might obscure problems.
2. Replace damaged components. They might be ok as-is, but why muddy the waters with unnecessary unknowns?
3. Replace parts that don't fit. Cramming parts in that don't fit creates the potential for mechanical stress that can crack solder joints and traces. Replace C17 with either a 1uF film cap that fits in the space available or use a 1uF tantalum (- end towards IC1).
Q5-Q7 are probably not making good contact in their "sockets" because the pins are too small and the fit is loose. It's really not necessary to socket BJTs. That cheap stamped-pin socket that IC1 sits in is unreliable. The only time I socket opamps is when I plan on "rolling" them, which is completely unnecessary in this circuit.
4. Measure the DC voltage on Q5-C and Q6-6. That will give use a clue as to whether this 1st & 2nd stage biasing is in the ballpark. Ideally, Q5-C & Q6-C should be equal to Vref. If they're within ±0.5V of Vref, it's good enough. I shoot for ±0.2V. The voltages should not change more than a few mV when the GAIN pot is rotated.
Once all that is done, we can regroup and plan the next steps.
